Fruit Trees and Family Trees in an Anthropogenic Forest

Home , Rattan

Fruit Trees and Family Trees in an Anthropogenic Forest: Ethics of Access, Property Zones, and Environmental Change in Indonesia Author(s): Nancy Lee Peluso Source: Comparative Studies in Society and History, Vol. 38, No. 3 (Jul., 1996), pp. 510-548 Published by: Cambridge University Press Stable URL: http://www.jstor.org/stable/179231 Accessed: 15-11-2017 17:38 UTC

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at http://about.jstor.org/terms

Cambridge University Press is collaborating with JSTOR to digitize, preserve and extend access to Comparative Studies in Society and History

This content downloaded from 173.250.174.7 on Wed, 15 Nov 2017 17:38:14 UTC All use subject to http://about.jstor.org/terms Fruit Trees and Family Trees in an Anthropogenic Forest: Ethics of Access, Property Zones, and Environmental Change in Indonesia

NANCY LEE PELUSO

Yale School of Forestry and Environmental Studies

Landscapes are culture before they are nature; constructs of the imagination projected onto wood and water and rock. . . . But once a certain idea of landscape, a myth, a vision, establishes itself in an actual place, it has a peculiar way of muddling categories, of making metaphors more real than their referents; of becoming, in fact, part of the scenery (Simon Schama 1995:61).

Setipa mountain rises behind Bagak Sahwa, on the south side of the paved trunk road between the city of Singkawang and the town of Bengkayang. On the other side of the road, a wide strip of irrigated rice fields meanders along a river that stretches north to meet the Selakau river. Bagak, a hamlet of more than 100 single-family houses, is laid out on a line lazily perpendicular to the trunk road, aiming towards Setipa's peak. The mountain and the hamlet are buffeted by a mile-wide strip of forest that winds around all the mountain's associated foothills. The village forest consists of tropical trees in mixed stands, scattered bamboo clusters, tangling vines slithering forth from the edges of cleared footpaths. The forest is dotted with only an occasional swidden patch of hill rice or a broad-leafed cluster of banana clumps. After the first half-mile, the tangles and thorns are left behind; the forest takes on a park-like quality. Open expanses under a layered canopy of tall trees make it easy and relatively cool to walk from one giant fruit tree to the next.

Many thanks to the friends who commented on earlier drafts of this essay: Jill Belsky, Louise Fortmann, Bruce Koppel, Christine Padoch, Jesse Ribot, and Peter Vandergeest. An earlier draft was presented at the Agrarian Studies seminar in January 1994. To the many colleagues who commented on and offered critiques of this article, I am truly grateful. I owe a special debt to Christine Padoch, as without her encouragement I would not have gone to West Kalimantan for this study. We spent many days in the field, in Pontianak, in New York, and on the phone, often over a durian or two, discussing the ideas in this essay. As is often the case, my ideas about Bagak have been heavily influenced by her ideas about managed forests at her field site and in general and by the people we met and talked with together throughout West Kalimantan.

0010-4175/96/3599-0884 $7.50 + .10 ? 1996 Society for Comparative Study of Society and History

510

This content downloaded from 173.250.174.7 on Wed, 15 Nov 2017 17:38:14 UTC All use subject to http://about.jstor.org/terms FRUIT TREES AND FAMILY TREES 511

Wrapping around the hillside, the park forest contains hundreds of three-, four-, and five-generation durian trees, many of them triple-armspreads around, the emergents in the tightly closed canopy. Because their planters have passed on, some trees are remembered now by the planters' names: Nek Bantang, Nek Limo, Nek Suhotn, Nek Rawoh, and Nek Garakng. Beside some of them are the trees planted by their living children and grandchildren: Si Anyap, Si Sulam. Other trees called Si were planted by men or women who died young-before receiving the honor of being called Nek (or grandparent). Some trees are named for the peculiar shape or taste of their fruit: Si Jongkup is named for its bottle-shaped fruits. A level below these trees in the multi-storied canopy are the progeny of fruit trees planted at the same time as the durian: tangy langsat, creamy angkaham, and sweet mangosteen. Up and down the slopes of Setipa's mountainside and the slopes of adjacent hills, these trees planted by Bagak hamlet's Salako ancestors have stood as witness to and subject of the village's history.

THE TRANSFORMATION OF PROPERTY AND ACCESS CONVENTIONS

In this essay I examine the transformation of property and access conventions in an anthropogenic forest in West Kalimantan, a province in Indonesian Borneo. Although both property rights and landscape are treated as processes which may react to transformations in one another, they cannot be unilinearly linked. This case illustrates the upredictability of local social and environmental changes resulting from state and market interventions which change property rights and landscape composition. The anthropogenic forest is examined in three aspects. One is the creation and management of forest landscapes by forest-dwelling peoples. I show how forest-dependent people not only deforest but also afforest or reforest sponta- neously, that is, without government or project sponsorship. This is percepti- ble only by looking at landscape change over a relatively long period-in this case, about 100 years. Examining only a moment in a landscape process, such as the burning of a forest patch to make a swidden or the extraction of products from the forest, obscures the larger management processes engaged in by local people. A second aspect of the anthropogenic forest is the notion that nature reserves created by governments or other outsiders can interrupt local forest management schemes and cause unanticipated changes outside reserve boundaries. Such interruptions, which ignore local management systems, may lead local people to find ways to alter the species composition of those spaces that the government now claims as a means of contesting sym- bolically their appropriation. A third but minor theme concerning the anthropogenic forest points out the ethnic differences in the manipulation of landscape components and the way in which one group's landscape forms can be appropriated by another. The occupation of a particular landscape by a differ-

1 The term nenek in the Salako language is used to refer to both grandmother and grandfather.

This content downloaded from 173.250.174.7 on Wed, 15 Nov 2017 17:38:14 UTC All use subject to http://about.jstor.org/terms 512 NANCY LEE PELUSO ent group may, in turn, lead them to reinterpret their processes of resource allocation and may unwittingly facilitate government access to the resources flowing from these landscapes. Focusing on property rights and access to forest resources, the examination of social relations in these forests pursues three themes. One theme explores the ways in which politics and discourse alter both the landscape itself and the means of accessing the resources in that landscape. To untangle these processes of change, I examine the ways that certain resources have represented or generated power, wealth, and meaning to those who control or have access to them (Shipton and Goheen 1992; Peters 1992). In particular, I look at how fluctuating power relations, production relations, and the social meaning of resources have generated different landscape relations between land and trees through multiple human generations occupying the forest. A second theme looks at the fluctuation across space and through time between individual and corporate (group) property relations in various resources, notably long-living trees. I call these fluctuations "temporal and spatial zones" of access rights and landscape management. The third theme analyzes the changing forms of resource access and what I call an "ethic of access" and the ways in which this ethic has, in some cases, tempered the potentially harsh consequences of broader (international) trends toward privatization, individualization, and commodification of resources. These social and ecological developments are not a unique artifact of a single village's history and experience. The managed forests of West Kali- mantan and their relationships to swidden and wet-rice production systems have only recently gained the attention of researchers (Ex 1992; Salafsky et al. 1993; Padoch 1994; Peluso and Padoch 1996). In many ways, they resem- ble the anthropogenic forests of East and South Kalimantan, which are dominated by cultivated rattan, rattan and rubber mixes, and other agroforestry production systems reported only in the past decade (see, for example, Lahjie and Seibert 1988; Potter 1987; Colfer 1993; Leaman et al. 1991; Peluso 1992c; Safran and Godoy 1993; Weinstuck 1983; Colfer and Soedjito 1995). Moreover, the managed forest ecosystems being created by these villagers have correlates in local-level forest management in other parts of Indonesia and the tropics in general (Alcorn 1981; Posey 1985; Denevan and Padoch 1987; Hecht, Anderson, and May 1988; Michon and Michon 1994; Sather 1990; Balee 1993; Leach 1995). The case, then, contributes to the growing scholarly literature on forest landscapes and the roles of local people in creating, maintaining, and altering them (Posey 1985; Hecht and Cockbur 1989; Padoch 1994; Schama 1995). I focus much of my discussion of resource tenure and meaning on durian (Durio zibethanus), placing this fruit tree in the context of villagers' management of other fruits, of rubber, and of agricultural land. Durian's value and demand all over the island of Borneo-throughout Southeast Asia in fact-

This content downloaded from 173.250.174.7 on Wed, 15 Nov 2017 17:38:14 UTC All use subject to http://about.jstor.org/terms FRUIT TREES AND FAMILY TREES 513

guarantees brisk sales wherever there are roads or rivers to transport it. Durian is an important component of both the social fabric and the physical landscape. The tree's biological characteristics of longevity and productivity affect property relations and illustrate the importance of temporal and spatial zones in the study of agrarian change. By scrutinizing changes in, and the contesta- tion of, property relations through the tree's many productive generations, we gain a bird's-eye view, as it were, of broader social changes in the landscape and their meanings. The following section explores resource tenure and concepts of temporal and spatial zoning as a means of understanding landscape change. I also define access and the ethic of access. Subsequent sections describe Bagak's past and present landscapes. The historical discussion elaborates transformations in the regional political ecology which have affected local resource management practices, including the property rights in various resources. I then discuss in detail inter-generational property rights and ethics of access for fruit, particularly durian. A shorter discussion of rubber and rice places durian and its social relations within the context of this particular landscape. How and why rights, access, politics, and discourse are changing is discussed in the following section. Finally, in the conclusion, I discuss the several ways that this case confounds some of the accepted wisdom of resource management, landscape formation, and property relations.

RESOURCE TENURE, ZONING, AND ENVIRONMENTAL CHANGE

We still barely understand the relationships between multiple forms of tenure in agrarian resources and the roles of property and political economy in shaping agrarian landscapes (Fortmann and Bruce 1988:11). We do know, however, that the explanation of environmental change is complicated by the interaction of such multiple aspects of property relations as, first, conflicting sources of legitimate authority, including customary and formal legal systems (Moore 1986; Fortmann 1990; Bromley 1991; Peluso 1992b, 1993); second, negotiated systems of meaning (Dove 1986; Posey 1989; Peluso 1992a; Ship- ton and Goheen 1992; Peters 1992); and, third, an individual's position in society, that is, their membership and relative autonomy (power) in different social networks (Blaikie 1985; Berry 1989; Okoth-Ogendo 1989). Political- economic institutions which influence access to resources such as markets, governments, and regulatory systems play an important role in the manner in which resources in different locales are used and managed (Moore 1986; Berry 1989; Bromley 1989, 1991). In other words, how resource access and control affect and are affected by power, wealth, and meaning will affect that resource's management (Peters 1992). Property relations in agrarian resources, like any social relations, are con- stantly shifting due to multiple influences and multiple negotiations (Moore 1986). A dynamic view of property compels a focus on process, rather than

This content downloaded from 173.250.174.7 on Wed, 15 Nov 2017 17:38:14 UTC All use subject to http://about.jstor.org/terms 514 NANCY LEE PELUSO

simply on the institutions, social structures, or bundles of rights and respon- sibilities that are particular outcomes of processes and negotiations. The study of property as process requires multi-layered analysis, as has been the practice in political ecology. That is, one needs to study both local histories and the layers of political-economic influences that affect local practice (Smith 1984; Blaikie 1985; Blaikie and Brookfield 1987; Neumann 1992; Peluso 1992c; Bryant 1992). As governments and policies change, as market opportunities or structure change, as the individuals or groups in power change, so do the "customary laws" which guide decision making and the practice of everyday life, including local resource management. In tropical forests, the complications of studying property relations and farmer decision making are confounded by the sheer number of species. Even if we limit our discussion to species that have economic, medicinal, or ritual value, the range and diversity of agroforestry products are immense, as are the tenure arrangements pertaining to them. To take a single example, there are many ways to carve up the bundle of rights to a tree, as Louise Fortmann (1985) has so aptly shown. The largest bundles of rights include those to own or inherit, to plant, to use, and to dispose of a tree. Each of these rights can be sub-divided; many of them interact with other factors. For example, tree tenure may determine or be determined by land tenure; or forms of access to these two resources may be independent. The right to use different parts of the tree may be allocated to different claimants, and these multiple uses may not always be mutually compatible. Land rights and tree rights may be held by different claimants, and change in one often, though not always, leads to change in the other. In addition, there are four broad classes of rights holders-the state, groups, households, and individuals-that may also be sub-divided into different kinds of groups (for example, by spatial units of residence, kinship, or legal association). Whether or not a tree was planted or self-sown can make a difference in the claims of rights holders; whether it is intended for subsistence or commercial use can also make a difference (Fort- mann 1985). One other influence on tree tenure bundles, to add to Fortmann's list, is a tree's natural characteristics: the length of time it grows and produces fruit, nuts, or other products; its reproductive strategies, that is, the number of seeds the mother tree produces; the manner of harvesting (whether the fruit is harvested or left in the forest); the survival rate of seedlings; and the intensity of management (Peters 1994). All these factors influence the tree's productive lifespan and people's ability to cultivate or to manage a self-sown tree. Bio- logical characteristics also play an important role in inheritance patterns. Date palms, for example, can be inherited indefinitely because new sprouts contin- ually emerge from mother trunks (Leach 1988). Natural characteristics often affect the value placed on the tree and the meaning of access to it. Long-living trees also teach a lesson in concepts of zoning. Zoning need

This content downloaded from 173.250.174.7 on Wed, 15 Nov 2017 17:38:14 UTC All use subject to http://about.jstor.org/terms FRUIT TREES AND FAMILY TREES 515 not be only a spatial category, in which separate spaces are set aside for crop production, reserve forests, economic forests, or other types of management. Zoning can also be a temporal categorization recognized by examining generational changes in rights to long-living fruit and nut-producing trees or self- propagating species such as the date palms mentioned above. Mixed-age stands of trees-a forest-can encompass three or more temporal zones of rights at one time. The differences between the zones can be expressed by changes in the rights to individual trees with each generation's passing. Over time, the composition of a landscape shifts with the zone of production and reproduction. Access to resources is often guided by an ethic of access (Peluso 1992a) which may be unique to the specific resource. An ethic of access is similar to the notion of a subsistence ethic (Scott 1976). However, for some resources, the ethic of access is driven by more than economics or subsistence rights and serves social, political, and ritual purposes as well, representing kinship, power relations, ritual harmony. Like other aspects of resource tenure, the ethic of access is a moment in a temporal zone of a larger property process-a dynamic concept rooted in both common and individual experiences of history but affected by a myriad of social, political, and environmental factors. At any moment in time, the ethics of access to particular resources are influenced by the physical characteristics of a resource, especially its longevity and divisibility; the type of use of a resource (whether for subsistence or commercial purposes); the existence of some social meaning for the resource beyond economic value; and changes in social relations affecting the balance between group or individual resource control, political-economic factors such as the prices of products and the relative ease of marketing them, and environmental circumstances such as a resource's relative scarcity, substitutability, or spatial competition with other products. As motivators for behavior and as indicators of change, ethics of access and their relations to actual practice are extremely important to understand for analyzing changes in resource management. In short, in this essay I will show that landscape as well as property are processes that respond to political-economic and cultural conjunctures. Responses to change are reflected in landscape composition and in the means of allocating access to resources, the value, meaning, and accessibility of which change over time. Access is used in a broader sense than property per se, although both resource access and property rights are key social relationships that influence-and are influenced by-the types of agrarian resources under consideration.

BAGAK'S LANDSCAPE TODAY Bagak is one of the two hamlets (dusun) in Bagak Sahwa village (desa). Located entirely on the south side of the main road, the hamlet is flush against the hillside forest which borders the Gunung Raya Pasi Nature Reserve. The

This content downloaded from 173.250.174.7 on Wed, 15 Nov 2017 17:38:14 UTC All use subject to http://about.jstor.org/terms 516 NANCY LEE PELUSO whole village has a population of approximately 1,096 people in 258 households; 114 of these fall within Bagak's jurisdiction.2 The hamlet is located in sub-district Tujuhbelas, in Sambas District, on a well-paved road running from the coastal city of Singkawang inland towards Bengkayang and beyond. Begun as a footpath when the area fell under the jurisdiction of the Sultan of Sambas, this road was widened and hardened in the 1920s and 1930s by the Dutch colonial government and paved by the Indonesian government in 1963. The original path led several kilometers past Bagak to the Sultan's gold mines in Montrado from the then small, predominantly Chinese, coastal settlement of Singkawang. Over the past eighty years this hamlet of Salako3 Dayaks has been infor- mally4 sedentarized by the encroachment of other land users on their borders and the fixing and formalization of these borders by successive colonial and national governments. The administrative village territory is bounded to the south by the relatively small (3,000-hectare) nature reserve created by the Dutch in 1932 as a watershed protection area for Singkawang. To the east of the village lie the agricultural lands of adjacent villages and special transmigration settlements for retired police and air force officers established in 1968 after the Confrontation with Malaysia5 and subsequent unrest in the area. To the west of the village is a Franciscan Catholic mission and school established in 1916. To the north are extensive rubber plantation lands established on the administrative village lands of Maya Sopa, which since 1980 have been worked by these villagers and some 4,000 Javanese transmigrants settled there. Bagak Sahwa today consists of some 2,700 hectares of land, according to official village statistics. This land was once heavily managed, farmed, and

2 I stayed in the village on six separate occasions between August 1990 and April 1996. The Community Forestry Unit at the Food and Agriculture Organization (FAO), Rome, funded the first two visits. Subsequent visits were funded by a travel grant from the College of Natural Resources, University of California Berkeley, while I was a Ciriacy-Wantrup postdoctoral fellow, a faculty start-up grant from the Yale School of Forestry and Environmental Studies, and a National Science Foundation grant (No. SBR-9310921), "Property, Resources, and the Globaliza- tion of Legal Systems." Most of the data reported here were collected in in-depth interviews and participant-observation. A survey conducted in February 1991 concentrated on a random sample of 40 out of the 114 households living in Bagak and recorded crop land, fruit, and managed forest holdings, as well as household income estimates. Sample households were revisited subsequently to discuss inheritance and other aspects of tree and land tenure. In December 1993, I made tree tenure transects in managed forests of various ages, collaborating with Dr. Charles Peters of the New York Botanical Garden for the collection of ecological data. Data on harvests and land use were collected by an assistant in 1994 and 1995. 3 Salako (the ethnic group) and Selakau (the river) are spelled differently in this article because of differences in practice. The Salako claim their origins lie along the Selakau River. On Indone- sian maps, the river is spelled Selakau. 4 That is, not by a specific government program. 5 The Confrontation was a period of conflict between Malaysia (mostly Sarawak) and Indo- nesia (mostly the provinces of Kalimantan), which lasted from approximately 1960 to 1965.

This content downloaded from 173.250.174.7 on Wed, 15 Nov 2017 17:38:14 UTC All use subject to http://about.jstor.org/terms FRUIT TREES AND FAMILY TREES 517 occupied by the ancestors of Bagak villagers, but now one-third (900 ha) is formally under the jurisdiction of the nature reserve, as part of the Ministry of Forestry's Department of Forest Protection and Nature Conservation (PHPA). Today, the law governing it requires villagers to acquire permits from the PHPA to enter the reserve, although this law is neither recognized by local people nor enforced by foresters. Foresters try to restrict certain forest uses such as swidden cultivation and hunting which, although rare, still occur. Foresters are lenient, however, about people collecting fruit from trees planted within the reserve's boundaries for reasons discussed below. This unofficial buffer zone on the border between the nature reserve and the village's fallow forests also provides an example of how local people have renegotiated their claims to the reserve. In addition to some farming of irrigated rice fields located in Sahwa, Bagak villagers practice a complex, long-rotation form of land management: a kind of cyclic swidden-fallow agroforestry (see Denevan and Padoch 1987; Padoch and Peters 1993). After the harvest of swidden field crops-rice, corn, cassava, and vegetables-and perhaps a second or third years' crop of peanuts or cassava, a swidden fallow is generally planted in rubber, fruit, or mixtures of rubber and fruit. Such gardens are managed subsequently for both the planted tree species and the self-sown trees which sprout in the interstices and are selectively protected during slash weeding (see Padoch and Peters 1993). Because of the long productive periods of the planted and self-sown trees (used, for example, for fuel or medicinal purposes), the swidden fallows become managed forests. Some swidden fallows are purposely left alone, without planting economic trees, allowing them to revert to successional forest in order to save the land for future field-crop cultivation. An alternative, or intermediary, land use is to plant a low-management but high production crop such as bananas. At any given time, each household manages numerous plots under different types of tree and field crops for consumption or sale. Until the introduction of rubber in the 1920s (see below), the landscape managed by Salako Dayaks consisted of largely swidden field crops and their multi-aged fallows, dotted with managed fruit forests planted in former living sites and some uncut or very mature hill dipterocarp forest. Mature forest not converted to swiddens or forest gardens was tapped for timber, resins, and other subsistence and commercial products. Rice self-sufficiency6 was crucial to local people when markets were few and food supply local; numerous large swiddens provided the family's food and dominated the landscape. In 1991, only a third of the villagers planted any swiddens at all; the average size of these was a meager one-third hectare. Economic trees now dominate the landscape. "Rubber," in the words of one local woman, "has become our

6 Or rice mixed with cassava and maize.

This content downloaded from 173.250.174.7 on Wed, 15 Nov 2017 17:38:14 UTC All use subject to http://about.jstor.org/terms 518 NANCY LEE PELUSO

daily rice"; and windfall profits accrue in good years to those with fruit to sell. It is likely that wet-rice fields will become more important in the next generation, but for now these extensive fields are only partially farmed. The precise amount of land under particular land uses is not clear, partly because some land uses change yearly, partly because government land-use categories are not consistent with local terms, and largely because statistics are not reliable. However, in 1990, the village head estimated that nearly half of the 1,800 hectares of village land outside the nature reserve were managed forests of one sort or another. More than 800 hectares of these managed forests contain mixed stands of rubber, durian, langsat (lansium domesticum), rambutan (Nephelium lappaceum), angkaham, cempedak (Artocarpus spp.), and other planted fruits, along with self-sown species used for timber, fuel- wood, and thatch. Self-sown rattan, bamboo, and medicinal plants are also protected. Of another 296 hectares of fallows set aside for planting in HYV rubber, 208 ha were planted between 1981 and 1988.7 The remaining cultivated land is largely in annual crops, with some 200 hectares of irrigated fields, although not everyone who owns these plots has enough time or labor to plant paddy rice. In August 1990, the village head estimated some 350 hectares were cultivated or fallowed swiddens.8 The remaining land not under the village's infrastructure, such as schools, playing fields, and housing, consists of relatively mature secondary growth which is not currently managed intensively.

BAGAK'S LANDSCAPE YESTERDAY: A SHORT HISTORY

In former days ... it was not the land that was important, but trees and their fruit. It was quite common for people to have [what is now called] "a garden" together, Mr. A with Mr. B, because they liked to plant their trees next to each other (Pak Manap, kepala adat, Bagak Sahwa).

The ancestors of the Bagak Dayaks lived in longhouses near the upper slopes of Setipa mountain inside and outside the borders of what is today the Gunung Raya Pasi nature reserve, in an area called Bagak Atas. Living in the moun- tains allowed them to diversify their production while defending themselves from their enemies. Not an aggressive headhunting group, these Salako did not make war in order to expand their resource territories, although they responded to war-parties' depredations with counter-attacks of their own. Villagers claim that their ancestors have occupied Setipa for several hundred

7 This is part of a program called Program Pembangunan Karet Rakyat (People's Rubber Development Program, or PPKR), in which smallholders are allocated one hectare or more of land and take loans of Rp. 4-6 million (US $2,000-3,000) to plant rubber trees that would yield within five years. This 296 hectares includes the more than 100 hectares that partially burned. These trees were largely planted as required by the World Bank as a monoculture plantation, in neat rows and columns 2.5 by 8 meters apart. (Some farmers, however, have innovated by planting the occasional pineapple plant or cempedak tree in the midst of their rubber.) 8 Since 1990, many of these fallows have been planted to fruit and rubber.

This content downloaded from 173.250.174.7 on Wed, 15 Nov 2017 17:38:14 UTC All use subject to http://about.jstor.org/terms FRUIT TREES AND FAMILY TREES 519 years. Until settling in their present site, the villagers moved their longhouse sites short distances, sometimes when the trees near the longhouse had grown too large, sometimes when illness struck the occupants. When they left a site, additional durian and other fruit trees were planted in the space left by the house structures. Such a site is henceforth called a timawokng (tembawang in Indonesian), meaning a former living site (see also Padoch 1994; Sather 1990). While the Dayaks practiced swidden cultivation in the dry hills, the swampy lowlands to the north were farmed by Hakka-speaking Chinese agri- culturalists. Their settlement village was called Sahwa and may have been established as early as the late eighteenth century, soon after Chinese began mining gold for the Sultan of Sambas at nearby Montrado (see Chew 1990). The Chinese cut the lowland forest and dug canals to drain the swamps and plant rice. Dutch colonial agrarian policy passed in 1870 and implemented in Dutch Borneo after 1874 prohibited Chinese and other non-native peoples from owning land. In thus dismissing the role of these peoples in converting this forest to productive irrigated agriculture, the Dutch ignored their own criteria in which land rights were recognized on the basis of continuous cultivation and the conversion of unproductive forest to useful crops. Their policy was legitimated by the creation of "native rights" and defining natives in ways that were not always relevant to the place. Chinese could only acquire usufruct rights to the land, through written agreements which were in effect long-term leases.9 Although they did not formally own the land, the Chinese continued to alter, improve, and make it productive. Chinese planters also leased rights from the Dutch colonial government to the drylands called Patengahan between the wet-rice fields and the mountaintops. There these Chinese planters built houses and eventually planted fruit and rubber. They were the first non-Europeans in the area to grow rubber, a crop with solely commercial value. The Bagak Salako generally maintained control of their forest and land resources, except for occasional tribute payments to the Malay Sultan of Sambas. The community sold or bartered agroforestry products with Malay traders or local Chinese shopkeepers who came to the longhouses seeking them; they also sold fish, illipe nuts (used for cooking oil at the time),10 savory-preserved durian (tempuyak), sweet durian cakes (lempok), and fresh fruit. Some longhouse dwellers walked to Singkawang, by-passing the Ma- lays, to sell directly to urban Chinese for higher prices. In 1920, the Dutch initiated plans to turn the upper slopes of the Gunung 9 These rights were called Hak Overencomst (or H.O.). 10 The people in this village talk about at least 3 kinds of illipe nut or "tengkawang," two planted for their fruits, one wild in the forest. The small one called bodot has a lot of oil. Other oil fruits include sang karengan which is a root with lots of oil, its fruit is like a gourd but is full of oil. Kelampei fruit also has good oil that tastes like peanut oil but does not itch (gatal), smells nice, and is a dark yellow/brown color. It is boiled then put out in the sun to dry.

This content downloaded from 173.250.174.7 on Wed, 15 Nov 2017 17:38:14 UTC All use subject to http://about.jstor.org/terms 520 NANCY LEE PELUSO

Raya Pasi complex, including Setipa mountain, into a nature reserve. The initial line for the reserve border, drawn where a water catchment device was to be constructed, encompassed all of the village's current and former living sites and their ancestral forests and lands as well. The longhouse occupants had to move off the mountain and forfeit these ancestral rights. Not surprisingly, people of Bagak Atas were not happy with the colonial government's plans. Dutch officials worked with local leaders to convince people that they had no choice but to move from Setipa to the lower slopes of Patengahan, near the areas that Chinese had planted rubber and fruit. 1 Even- tually people moved out of longhouses into single-family dwellings built large enough to accommodate extended families. The villagers' unhappiness with the new spatial arrangements caused some families to remain on the mountain (inside the reserve) until approximately 1940, eight years after the nature reserve was officially established and mapped. These families moved only because the controlleur jailed some of them for several weeks when they refused to leave or continued to make swiddens above the reserve boundaries. The head of the new village continued to negotiate with the Dutch to recognize the rights of local people who had planted trees or converted forest there by changing the reserve's boundaries. Though it is not clear when they did so, the Dutch eventually moved the reserve boundary to a place above the old longhouse sites, thereby restoring a good deal of the people's territory; this constituted a victory for the local people. The villagers have continued to push back the border in less-organized, informal ways. Villagers harvested rubber, durian, langsat, rambutan, angkaham, cempedak and other fruits planted by their ancestors within the border. During the Japanese occupation (1942 to 1945), the Indonesian revolution, and the early years of Indonesian independence, villagers made swiddens within the reserve boundaries and planted rubber and fruit in the fallows. Government surveillance of state forests was practically non-existent at that time. By planting productive tree crops, people were also staking their claims to their control of the hillside's upper slopes by negotiating new forms of old territorial and resource claims. The border continues to be blurred, since people still plant fruit and rubber trees within the reserve, thus creating an informal buffer zone between the reserve forest and Bagak village forest. They also hunt an occasional deer and collect the products of various self- sown species such as bamboo shoots, candlenut, and rattan within the borders of the reserve. And, as late as 1976, people were able to move the border back another 100 meters when they were working on the construction of the permanent border markers. The intervention of the colonial state in the arrangement of the village's

11 The name Patengahan still appears on Dutch maps from the 1930s.

This content downloaded from 173.250.174.7 on Wed, 15 Nov 2017 17:38:14 UTC All use subject to http://about.jstor.org/terms FRUIT TREES AND FAMILY TREES 521 social and productive space did more than superimpose a new set of rules about land and forest use. Real and imaginary boundaries were imposed for the first time on the villagers' access to the lands and land uses they had always preferred.12 Current and past events were converging to create pressures for an entirely new landscape. The change in spatial arrangements helped spark Salako interest in planting rubber. After people were denied legal access to the extensive tracts of reserve and secondary old growth on the mountain, it became difficult for everyone to find enough swidden land to produce their families' food. Some people, nearly half of the old longhouse's population, moved north of Sahwa and established a new village near swidden fallows they had farmed in the past. A few of the Dayaks who stayed experimented with planting irrigated rice near the Chinese but gave up-the time and labor demands of wet-rice production interfered too much with their other productive activities. Rubber was much less labor-intensive, and Chinese planters were doing well. By the 1930s, people could exchange rubber for government-issued coupons to buy food and other supplies on credit. After moving out of the reserve and thus losing access to land for agricultural production and after the price of rubber began to rise, the villagers' interest in rubber began to increase. Rubber's biological characteristics also made it an attractive crop. Latex could be collected and marketed in virtually all seasons except the rainiest weeks, and uncollected latex did no harm to the tree's productivity. Until the Confrontation, latex could be carried (without spoilage) to Malaysia and sold or exchanged for goods there. Latex could also be sold along the Singkawang road, which the Dutch widened in the 1930s. Rubber trees increased the value of the villagers' land and increased a household's diversity of economic options. When the first Dayaks moved down from Setipa, none planted rubber yet. Gradually, it became popular; and by 1991, 85 percent of the villagers owned productive rubber trees. 13 As late as the mid-1960s, however, Chinese rubber production still dominated the local market. The current village head claims that when he arrived in 1965, Chinese smallholders were producing most of the one tonne of rubber exported daily from Bagak. As the village's resource territory shrank, the relative intensity of land use became increasingly important. As a result, the combined impact of the loss

12 To illustrate the contrast with previous interventions, the Chinese had moved into an ecosystem that was not at the time exploited by the mountain people of Bagak. When the Malays exacted tribute in the form of forest products, they did not move into the village and take over the territory within which these products grew. Even the missionaries were not interested in evicting local people from the lands that they were most psychologically and physically dependent on. For the first time, these Dayaks were squeezed between state-appropriated and controlled land and Chinese cultivators in the irrigated lowlands. The Dutch planned to let the watershed go back to undisturbed forest; the Chinese had already converted the swamp forest to wet-rice production. 13 This contradicts a study by Dove (1993) who maintains that swidden production was preferred over rubber production as a means of resisting external control.

This content downloaded from 173.250.174.7 on Wed, 15 Nov 2017 17:38:14 UTC All use subject to http://about.jstor.org/terms 522 NANCY LEE PELUSO of land for expansion and the response to market opportunities in rubber planting, by the mid-1960s, many people were no longer able to produce their year's supply of rice on their own land and had to buy rice. Rubber trees in particular began encroaching on swidden fallows. To feed their families, then, the villagers had to sell more of their forest and agroforestry products. These stresses on land and changes in production perhaps contributed to the local people's willingness to participate in evicting the Chinese from the lands that they and their ancestors had transformed and managed for some 200 years, in spite of the strongly held belief that labor invested in land management and forest conversion imparted inheritable rights. In 1968, the Indone- sian army, assisted by local Dayaks, violently forced the Chinese to move from the agricultural areas to urban centers such as Sambas, Singkawang, and Pontianak.'4 The evicted Chinese left behind approximately 150 hectares of paddy lands, plus productive rubber and fruit gardens. Important new changes in the landscape and the distribution of formal rights of resource control began to emerge at this time-once again largely as the result of the ways that the state intervened in allocating, formalizing, and enforcing property rights in land. First, when it became clear that Chinese farmers would not be allowed to reclaim their land, a council was organized by civil and military officials from the subdistrict. This council, which included the village head, the head of customary law, and some informal village leaders, divided the irrigated fields and fruit and rubber gardens that the Chinese had cultivated amongst local Dayaks. Some people had posted claims to particular parcels before the meeting was held; and the council formalized their claims. Other people refused to take any of this land, feeling that "the land was filled with tears," as one villager put it. Claims were made official at the agrarian office, where people paid nominal registration fees for their land certificates. Second, the territories of Patengahan and Sahwa were combined into a single village called Bagak Sahwa. Not long afterward, the village's physical borders were sealed much tighter than they had been when the administrative villages had been established, as transmigration areas for retired police and retired air force personnel were set up on the eastern border of the village.15 Besides rewarding these military men for their service to the country, the state was able to attach symbolic meaning to its role in enforcing permanent settlement patterns. Despite the presence of former military families, additional conflicts over lands and resources erupted in the late 1970s

14 The complexities surrounding social and political relations of this violent period and the circumstances around the exile of the Chinese have been difficult to substantiate by research from field work and oral histories alone. Moreover, research on ethnic conflict of any sort is strongly discouraged and forbidden by the Indonesian government. Without having access to archives from this period, it is impossible to tell the nuanced story that this tumultuous period deserves. 15 Plans were eventually made to settle some 4,000 Javanese families in a transmigration settlement on the lands of Maya Sopa.

This content downloaded from 173.250.174.7 on Wed, 15 Nov 2017 17:38:14 UTC All use subject to http://about.jstor.org/terms FRUIT TREES AND FAMILY TREES 523 between local people and recent migrants from Madura. Again, violent incidents led to the literal eviction of many newly settled Madurese.16 The early 1980s saw a second surge in rubber planting, when a government project to fund smallholder production of high-yielding rubber (PPKR) was introduced. The state's role in this project changed its relationship to the nature of village land rights. Land converted to high-yielding rubber production changed its status from "customary" village land (tanah adat) to state land (tanah negara), while people gained new rights to the high-yielding rubber trees. The bank now held the title to these lands, and once the loan for the new production system was paid, the title went to the villager who paid it. Nevertheless, the state is the ultimate guarantor of such titles, giving itself a new property right in the land that it had not had when the land was considered tanah adat. Thus, while landed property was increasingly important to the ownership of trees in village forests, individually owned trees were gaining importance on lands converted to state property. 17 Once the loans are paid off, the owner receives a land title. Rubber was now both a means by which people put relatively long-term claims on a piece of land and a way for the state both to reclaim land from the villagers temporarily and to retain some rights over it permanently as the authorizer of title. In addition to the changes brought on by rubber, the land use revolution has also entailed an expansion of land use categories and a change in the types of places that fruit trees are planted. Easy access to the urban markets of Singkawang and Pontianak, facilitated by the paving of the road in the 1960s and the entry of Japanese vehicles (motorcycles and vans) in the 1970s, added additional incentives for planting trees. Durian trees are no longer found only in former living sites but are taking over swidden fallows and invading rubber gardens (which were themselves planted in swidden fallows). By 1991, some 97 percent of sample households had planted durian trees in swidden fallows; at least 41 percent had planted durian in or just next to their rubber gardens. Durian stems occupy 71 percent of the total basal area of old and medium- aged fruit forests (Charles Peters, personal communication, 1994).18 The increasing ecological and economic importance of trees in the landscape has increased the value of the land on which they are planted. Fruit trees

16 Although the three incidents leading to violence between Madurese and Dayaks are little known outside the province, the memory of them within West Kalimantan is still quite fresh. A monument was erected at the site of the peace agreement memorial by the Indonesian Army along the Singkawaing Bengkagang road and at the entrance to Samalantan, where much of the activity took place. In Bagak Sahwa, it seems that ten to fourteen families were evicted, many of whom had bought, borrowed, or occupied the ill-fated land left behind by the Chinese only six to eight years earlier. 17 Although in reality, the bank (People's Bank of Indonesia, or BRI) owned the trees until the loans were paid off-a legality that benefited local people when more than half the trees burned in the 1984 fire. 18 Basal area is the area of a cross-section of trees at breast height.

This content downloaded from 173.250.174.7 on Wed, 15 Nov 2017 17:38:14 UTC All use subject to http://about.jstor.org/terms 524 NANCY LEE PELUSO

had long been the property either of an individual or a descent group; in addition, clearing forest for swiddens always gave the person doing the clearing household rights to that land in perpetuity.19 Although it was not uncommon in the past for someone to plant a cluster of three to six fruit trees (kompokng) and thus to occupy a piece of land, the notion that a garden (kabotn) consisting of a relatively large unit of land would be valuable because fruit trees are planted there is a new concept. In sum, the relationships between land, trees, and the systems of tenure around them have been revolutionized by major changes in the region's political ecology. These have included: the sedentarization of village borders; the fluctuating enforcement of the nature reserve's borders, and with it the fluctuating access of local people to the land and forest resources within those borders since 1932; the introduction of two forms of rubber at two different times; the eviction of the Chinese and the appropriation of some 150 hectares of irrigated land for wet-rice production; and the booming market for fruit brought on by political economic change and the improvement and expansion of transport facilities. Each of these changes either modified the villagers' access to resources or expanded the markets for local products. The outcome of these changes was reflected in the landscape at different historical moments in shifting spatial zones of both forests in relation to fields and rights and mechanisms of access to these. The transformations also altered the ways some people valued the mixture of resources and led to new types of property relations in many of them.

FRUIT TREES AND FAMILY TREES

The previous section described the ways that particular historical events of local, regional, national, and international importance shaped the current landscape in Bagak. In other words, we have been looking at the ways that human processes and events have shaped the landscape. If we think of spatial zoning as having a temporal dimension, we have simply observed how forest became fields and then a new type of forest in what I call a landscape process. We have also seen how, at a macro level, property rights and access to the various resources created in those spatial zones were shifted, pushed, and pulled by actors with varying degrees of power and authority to do so at different points in time. The second half of this essay focuses on the themes of zoning and ethics of access as these are experienced in the lives of different people, as philosophy or discourse, and embedded in the treatment of particular resources of importance to local people. The ethic of access is discussed as it plays out in the

19 Land rights for forest clearance and swidden cultivation are well documented for Borneo; see Geddes (1954), Freeman (1955), Appell (1970), Weinstock (1983), Padoch (1983), Dove (1985, 1988), Rousseau (1990), and others.

This content downloaded from 173.250.174.7 on Wed, 15 Nov 2017 17:38:14 UTC All use subject to http://about.jstor.org/terms FRUIT TREES AND FAMILY TREES 525

system of inheritance, with an emphasis on the inheritance of tree crops. Another aspect of the ethic of access is its influence on the zoning of resource control and access through space and time. Here, the differences in the biological characteristics of specific trees and field crops play an important role in the extent, intensity, or duration of zoning, thus illustrating how ecological and sociological processes interact within a landscape.

Inheritance Ethics

For resident villages in Bagak, land and trees still provide the most important sources of household income. Rights of resource access and control are recognized by the community on two bases: kinship-the bilateral inheritance of rights to various resources-and the investment of labor in resource production or management.20 In the case of most fruits and forest products, partici- pation in the harvest is both a right by inheritance and a conveyor of rights by dint of labor investment. Some examples of the ways that labor imparts preferred rights to one member of a group of inheritors include clearing fruit or rubber gardens for a swidden, planting a tree on a piece of commonly held land, or clearing the brush from around a fruit or rubber tree to facilitate harvest of the tree's products. In principle, both male and female children share equal rights to inherit their parents' resources, although primary rights to a household's resources remain within the household unit. The household, or the husband and wife plus resident children and elderly parents, manages a set of land-based resources in addition to any cash savings and some of the movable property acquired by household members working outside the village or outside of agriculture. The cluster of land-based resources includes inherited lands under cultivation and in fallow, inherited durian and other fruit trees, rubber gardens, and lands cleared and trees planted by the parents during their lifetimes. When people lived in longhouses, the household unit was defined as an apartment (bi'ik) in the longhouse. After they married, children generally moved out of the apartment and built their own place onto the parents' longhouse or another one in the same general settlement area. Whichever child remained within the apartment to care for the parents in their old age would inherit primary rights to the household's total resources. This child's spouse would move into the apartment; after having children, the pair would eventually take over as co-heads of household. When the parents died, the child who had lived with, and cared for, them generally managed the household's resources and divided the inheritance among his or her siblings. A larger share

20 Fortmann (1988:20) refers to this as "the doctrine of labor creates rights"; Locke also saw labor as the critical motivator and confirmation of property rights. Interestingly enough, this "doctrine" seems to hold true in both Western and Eastern societies and has correlates in both ancient and contemporary property systems.

This content downloaded from 173.250.174.7 on Wed, 15 Nov 2017 17:38:14 UTC All use subject to http://about.jstor.org/terms 526 NANCY LEE PELUSO generally went to this person in exchange for their financial outlay in caring for the parents.21 Although most nuclear families live today in single-family homes, the concept of primary rights going to the child who resides with and cares for the parents in their old age remains. As discussed below, however, primary rights are not supposed to exclude other siblings from gaining access to the productive resources of their parents and grandparents. Perhaps for this reason, local people refer to what Western analysts call the primary rights-holder simply as "the person in charge" (yang kuasa). The term implies that other kin have rights which are supposed to be administered, but not monopolized, by the one in charge. The person in charge of the inherited trees may be a male or female child and may be the eldest or youngest, the youngest being the most common among the Salako simply because they are the last to marry. Several points need to be made to explain the impact of this inheritance system on the distribution of resources of the village or descent group. First, because inheritance is bilateral and marriage is not restricted by class or status group qualifications, descent groups overlap. Bilateral inheritance means that an individual is a member of many different descent groups at the same time and has access rights to the trees planted by ancestors common to others in the descent group. Each husband-wife team is a member of two to four descent groups holding grandparent trees, depending on whether their grandparents planted trees as a couple or individually. They also have claims to grandparent trees planted by great and great-great grandparents. Thus, a child of parents A and B has rights to the fruit of trees planted by both the maternal and paternal grandparents. He or she also has rights to the trees planted by the grandparents' parents and grandparents. Again, because inheritance is bilateral, the descent group formed by common descendants of each grandparent or great grandparent will be different. This membership in multiple descent groups hypothetically prevents one or several descent groups from dominating or monopolizing land and other resources such as long-living trees. At marriage, the couple's families decide which family they will live with or near and inherit from. The system revolves around the availability of trees and land to a descent group, which is directly related to the ancestor's invested labor in clearing forest and planting trees. The second point about resource distribution is that part of the inheritance principle imparts greater rights to one of the children in a household-the one

21 This pattern is common among many Dayak subgroups throughout Kalimantan and Sar- awak: Whichever child takes care of the parents until their deaths generally inherits the primary rights to control the swidden lands that the parents cleared and the cash crop trees they had planted or protected. In some Borneo societies, however, land and other resources are not passed through an inter-generational household unit but are returned to the community's common pool when the household heads (husband and wife) pass away (Appell 1970). Rubber trees do not usually last through more than two generations and therefore are not subject to the complex inheritance rights associated with fruit trees, especially durian.

This content downloaded from 173.250.174.7 on Wed, 15 Nov 2017 17:38:14 UTC All use subject to http://about.jstor.org/terms FRUIT TREES AND FAMILY TREES 527 who continues the household and cares for aging parents. This individual gains the right to manage or take care of (ngurus) the household's resources, which generally means deciding how access to these will be allocated. As described below, this may entail rotating access to fruiting trees on a seasonal, weekly, daily or other basis or may involve temporarily or permanently dividing up the rights to particular trees among living siblings. Thus, we see several disparities between principles and practices. Both the rules of intergenerational inheritance and of the ethic of access articulate a relatively egalitarian mode of resource access. In practice, however, having one member of a descent group being in charge of the tree, even though this person will change through the generations, gives him or her greater power over the resource and a greater control over everyone's access. On occasion, while custom dictates that grandparent trees have a different inheritance status than the trees planted by parents (see below), people speak of unusual cases in which the primary rights to grandparent trees are not acessible to the group but are taken over by a single sibling. Even in such cases, however, the individualization of control over these trees does not diminish the ethic of access to them-other children and grandchildren retain a claim to the fruit, and the person in charge is expected to allow them to collect fruit for a night or two if that claim is exercised. The data on inheritance transactions show that women are somewhat less frequently in charge of family trees than men. In a survey of inheritance transactions of all the trees in two of the oldest tembawangs (Timawokng Anjauh and Gantekng Raru) and across a transect through three medium-aged tembawangs (Timawokng Rumoh, Paguk Batu, and Sirarok Atas), out of 183 inheritance transactions, 38 percent of the time primary rights were descended to female children and 62 percent of the time to male children or nephews. Thus, there seems to be a slight male bias in inheriting the status of being "in charge." Local people explained this by saying that most women did not formerly manage grandparent trees in old tembawang out of a fear of head- hunters. This practice reflected local experience of three generations earlier. On the other hand, a calculation of the choice of residence of newly married couples showed that the sample of forty households was evenly divided between living on the land of the wife's or the husband's family. Other social mechanisms exist to prevent the monopolization of inherited resources by a single heir. For example, although the child who cares for aging parents in principle has primary rights to the household's land-based resources, parents sometimes divide up the trees before they die in order to prevent discord among the siblings or to ensure that a favored grandchild or even a niece or nephew gets a specific share. In some cases, the children agree to maintain equal access to the trees and share relatively equally in the trees' management and fruit harvest. In other cases, some children of the tree planter are effectively eliminated from their inherited shares by sibling discord, by moving so far away that travel to collect fruit or cultivate a swidden

This content downloaded from 173.250.174.7 on Wed, 15 Nov 2017 17:38:14 UTC All use subject to http://about.jstor.org/terms 528 NANCY LEE PELUSO

is impractical, or by a sibling withdrawing from the co-heirs' pool. Finally, it needs to be noted that the ethic of access mentioned above and described below strongly discourages the person in charge from monopolizing access to all the family's trees.

Zoning, Temporal and Spatial The most complicated of the three landscape components, fruit is becoming even more so, as regional political ecology and property relations change. Many fruit trees produce fruit through three to five human generations, but durian may produce through seven. With each generation, the ideal rules of access change. In other words, the predominant rule of access to fruit changes from being held individually to that of being commonly held by an increasingly larger group of descendants, although some descendants have more rights than others. As the generations pass, more and more people have inherited rights to the tree's fruit. Particularly noteworthy is that people insist that moving their residence does not necessarily eliminate one's rights of access to ancestral fruit trees; it is not uncommon for children or grandchildren to return to the village during the durian season to share fruit with their siblings or other kin (compare, Appell 1970). Sometimes co-heirs who live in one village will seek out a particular grandfather tree located in another village and are rarely turned down in their request for a few days of harvest rights from the person in charge. These kinds of journeys, however, seem to be relatively rare and are more a matter of curiosity about ancestry than an occasion to demand a share of the harvest. The temporal or inter-generational zoning of rights is generally interpreted as follows: While alive, the tree planter has exclusive rights to the fruit, although it is usually consumed with other household members and often with guests and family members in other households. Profits from the sale of any fruit belong to the household or the planter. After one spouse passes away, the other is in charge of allocating access to the fruit, again generally sharing the annual harvest with other members of the household and sometimes the immediate family. After the tree planter and spouse have passed away, all the male and female children inherit rights of access to the tree's fruit. However, as mentioned above, the child who continues the household usually retains primary rights, in that he or she makes the final decision as to who will have access to the fruit among the siblings. In the third generation, all the grandchildren of the planter-a co-heir's group of cousins-have equal rights in principle. Once tree rights pass to the third generation, the tree is called a panene'an (grandparent tree), and in the past the claims of the total group of descendants allegedly superseded the claims of the person in charge of the tree. Many local informants insist that rights to the fruit of panene'an trees were not to be monopolized by an individual. The same rule held true for the fourth, fifth, and sixth successive generations of tree holders. One reason for the difference

This content downloaded from 173.250.174.7 on Wed, 15 Nov 2017 17:38:14 UTC All use subject to http://about.jstor.org/terms FRUIT TREES AND FAMILY TREES 529 between the children's generation and those of the subsequent grandchildren is the fact that the debts of labor and financial investment in parental care had been settled by favored resource access in the children's generation. Each villager is in various ownership categories or temporal zones of resource control at any point in time. Tree planters (and their spouses) are exclusive owners; the children of the tree planters are small group owners (co- heirs); grandchildren and great-grandchildren constitute larger co-heir groups as the generations pass. At different life-cycle stages, a person's access to resources tends to be dominated by different property zones. As individuals or young married couples, they still depend on the efforts of parents and grandparents, in some cases inheriting part of their parents' productions and in many cases having access to grandparent trees. When they still live with their parents, they share the parents' share. As people grow older, they tend to harvest fruit primarily from the trees that they planted themselves. This last point illustrates another aspect of the ethic of access. Each contemporary mature generation has a responsibility to their younger and future generations, that is, to plant enough trees to enable them to first relinquish their use of the grandparent trees (but not their rights in them) and then to create the grandparent trees of the future. As noted earlier, even if they have pulled themselves out of the descent group pool, people may still join in a few days of the harvest of favorite trees to enjoy a particularly delicious fruit, to sample a variety of durian "flavors" in a season, or simply to have the fun of gathering with one's cousins. How well one's ancestors fulfilled their planting obligation is reflected in current resource distribution, landscape composition, and the number of trees or sites to which one claims access rights. How access is divided among co-heirs differs according to the species, the tree's natural characteristics (especially those related to productivity), and its value. Some fruits, such as rambutan, mangosteen, or rambai, are freely picked and eaten by village children.22 The fruit from trees which must be climbed and for which there is a good market is generally divided equally among all claimants, with only sometimes an extra share for the tree climber. This is the case for rambutan, langsat, angkaham, and coconut. For some families, these picked fruits may be distributed among the co-heirs, whether or not they come to the harvest; others require a physical presence when the tree is climbed. Co-heirs who had moved out of the village may or may not be notified of a picking date and take part in the harvest, or they may have relinquished their claims. Again, this varies across families.

The Case of Durian Durian is a special case, again, partly because of its physical characteristics and partly because of its social meaning and economic value. When durian

22 In this, children can be said to have specific access rights to fruit, not unlike those described by Aschmann (1988) in speaking of a "child's culture" of untrammeled claims by children of southern California in the 1930s to privately owned fruit growing in people's yards.

This content downloaded from 173.250.174.7 on Wed, 15 Nov 2017 17:38:14 UTC All use subject to http://about.jstor.org/terms 530 NANCY LEE PELUSO reaches its peak of flavor, it drops from the tree. At that point, it must be eaten, processed, or sold, because as the days pass, it rapidly loses flavor, quality, and value-all indicated largely by the intensity of the fruit's fra- grance. The fruit of a single tree do not drop all at once but over a period of approximately seven to twenty-one days. Nor do all trees ripen at once. Durian season in a village may last two months or more. In Bagak, as in many other villages, durian fruits are never plucked from the trees. The head of custom (kepala adat) in Bagak says that people who plucked their durian used to be fined because this affected the entire village by tarnishing the reputation of Bagak durian in the marketplaces of Singkawang and elsewhere. The uncertainty of a tree's production schedule, the necessity for immediate harvest, and the trees' longevity have led to a number of interesting harvesting rules and practices applicable to the co-heirs' group. First and most important, to exercise a claim to inherited durian, a co-heir must be present at the harvest, that is, when the fruit drops from the tree. This entails building a temporary shelter in which to wait for several weeks and carrying the fruit by the basketful down the hillside or selling fruits to carriers who make daily rounds through the forest gardens. Before the durian harvest, co-heirs may have participated in other forms of labor, usually by clearing brush from the base of the tree where the fruits were to drop, to seal a claim to a harvest share.23 The number of trees that a descent group holds also affects the allocation of access and whether or not spatial or temporal zoning of access is important. When an ancestor planted only a few trees, the descendants are likely to rotate access rights in a system called baboros. Baboros has diverse forms: Inter- ested rights-holders may each take turns waiting for a few days, divide the waiting time between days for some and nights for others, take turns over different seasons, or all wait together and divide the day's or night's take equally amongst themselves. Baboros is a short-term temporal zoning strategy. Each claimant also has the right to send a proxy-a child, a nephew or niece, or a person with whom they agree to split their share. When an ancestor planted many trees in many places, it is more likely that the children or grandchildren will divide up access rights to the fruit by trees or clusters, using a spatial zoning strategy, as it were. Especially when trees were planted far from each other, it may be physically impossible for the whole group to harvest all the falling fruit together. Temporal and spatial zoning strategies may be combined by the agreement of family members to allocate exclusive access to trees to individuals for several years or the duration of the generation. Durian has a special ethic of access attached to it. Unlike rubber, durian is an ancient crop. Although it is not truly native, durian has long been natu-

23 See Peluso and Padoch (1996).

This content downloaded from 173.250.174.7 on Wed, 15 Nov 2017 17:38:14 UTC All use subject to http://about.jstor.org/terms FRUIT TREES AND FAMILY TREES 531

ralized and distributed throughout Borneo (Ashton, personal communication; Padoch and Peters 1993).24 Durian trees are markers of human settlement: Wherever durian trees are found in the forest, settlements have existed at one time or another (Padoch and Peters 1993; Sathers 1990). Although many fruit trees may be planted in the gaps left by a longhouse which has been moved, durian trees are always planted; they are something of an indication of a tembawang. Because they not only mark the location of a village's current but previous series of sites, durian trees can provide a history of a village's movements.25 Village social history is connected to its old durian trees in other ways as well. Unlike other trees, durians are named after the planter, after the flavor or textural qualities of the durian itself, or after an unusual geological feature of the landscape near the durian tree. Frequently the tree is named after an event-something that occurred near the durian tree-or an action taken by someone near that tree. For example, at a place where an individual dug water from a spring on a day when people working nearby were thirsty, a durian tree is named after the digger or after the thirsty people. A durian tree where two brothers fought over the division of the fruits, one throwing his share down the hill in disgust, is named for this event. Social history-the remembrance of events-is thus inscribed in the landscape. As one farmer said, "We couldn't write their names in a book so we remember them by the trees they planted." Thus, whether or not they have inherited rights in particular trees, villagers share a collective interest in these trees-a claim to the history represented by the durians' presence in the landscape. An interesting aspect of the naming process is that a tree can be named only after it has begun to bear fruit. This emphasis on the bearing of fruit mirrors other social customs-for example, that a person is not fully considered a person until he or she has children and that a child and his or her spouse can become heads of the continuing household only after having children of their own.26 When a person has passed on, the tree or cluster of trees he or she has planted is named after them but with different honorifics attached, depending on the planter's status as a parent or not when they died (the term Nek refers to a parent or grandparent, while Si is used for someone who died young without having children). For the descendants of the planter, the durian harvest is a time in which one recognizes one's family. Because the ripening characteristics of durian require the harvesters to wait for the fruits to fall, because the trees endure through

24 According to Peter Ashton, durian has its origins in Thailand but has been naturalized in the Malay archipelago for probably thousands of years. 25 Peter Brosius (n.d.) makes a similar point for the Penan of Sarawak, whose conception of the landscape revolves around the social-historical events that took place along rivers, forest paths between rivers, at the base of certain trees, and so on. 26 Of course, in practice this may be altered, say, if the child caring for elderly parents is unable to have children and if no other children are able or willing to take on that role.

This content downloaded from 173.250.174.7 on Wed, 15 Nov 2017 17:38:14 UTC All use subject to http://about.jstor.org/terms 532 NANCY LEE PELUSO numerous generations, and because the harvest experience has meaning, these trees are more enduring representations of kinship than other inherited resources.27 All relevant family members have a recognized claim to co-owned fruit; to refuse a sibling or a cousin access for consumption is considered very bad form and likely bad luck (the trees might not fruit ever again). Groups of siblings and cousins or households of young married couples and young children wait together to share the fallen fruits on the spot or divide them up for sale. Even though a great deal of durian is sold today, much is consumed when and where it is best: at the moment and in the setting where it drops. Any durian opened for consumption will be joyfully shared by all present, including passers-by and guests of the co-heirs, all of whom feel free to comment critically on the texture, taste, or quality of the fruit. As falling durian fruits know no time of day, waiting in the pondok at night has its own special thrills. The oldest trees are located in old longhouse sites, and harvesters often see or hear spirits in the forest while they wait. Few young people are willing to stay alone in a pondok at night. The hillsides are noisily occupied during the one to two months of durian season. The oldest durian trees in Bagak, apparently planted by another Salako group before the ancestors of the current residents came to the site, grow in the vicinity of the warriors' burial grounds. Each one of these trees bears a name, but villagers cannot explain their meanings because the stories disap- peared with the people whose ancestors planted them. These trees, people say, are owned by the whole village and owned by no one person. Even people with gardens adjacent to these trees will not claim them as their own-they are publicly recognized as open access trees. When their ancestors first occupied this territory, the durian trees were part of the landscape and provided the villagers with fruit. Because the trees provided their ancestors with fruit without the need to invest labor in planting or maintenance, these fruits have been slated ever since for residents or visitors who had no access to durian in season because their trees were not fruiting that year, because they were newcomers, because they were just passing through, or for other reasons. The trees were, thus, part of the village's common heritage. So strong are the feelings about the role of these trees in the village's social history that even those which stopped producing and were subsequently cut down are still talked about in the present tense. Other customs render even owned durian trees an open access resource under specific circumstances. For example, anyone can eat another's un- guarded durian which happens to fall just at the moment he or she is passing

27 Illipe nut trees in some areas also represent common ancestry to co-heirs but because most trees only fruit in cycles of four years or more and because the nuts are not consumed en masse and on the spot, the experience (and the memories) are qualitatively different. Illipe nut trees are not named. The low prices of illipe nuts, the uncertainty of their markets, and the usefulness of their wood for construction have caused many Bagak villagers to cut them down. Few illipe nut trees remain on village lands or in the buffer zone between the village and the reserve.

This content downloaded from 173.250.174.7 on Wed, 15 Nov 2017 17:38:14 UTC All use subject to http://about.jstor.org/terms FRUIT TREES AND FAMILY TREES 533 by, whether or not the lucky person has inheritance rights to it. Similarly, a group of co-heirs waiting for their durian always invites a passer-by to share in fruit which falls while the latter is passing. Not accepting such an offer or failing to follow ritual procedure for refusing it can lead to fatal accidents, especially poisonous snake bites, cutting oneself with a bush knife, or falling out of fruit trees.28 Because of the durian tree's longevity and its complicated inheritance patterns, creating durian gardens imposes new constraints on the farmers' land- use decisions and could become a means of changing the distribution of village resources in land. Land under trees which live for four to six generations is taken out of field-crop production for a significant period-a century or more. Within two generations of a tree's planting, a group of people with co-heir rights have a right to prevent one of the group's felling the durian trees for individual benefit (for wood or for access to the land underneath). Some not-so-obvious benefits may derive from this situation. First, converting land under rubber or rice to durian may redistribute benefits (after at least a generation) amongst many people rather than to an individual or a single household. Those common rights in the resource and the space it occupies can extend through many years-a considerable temporal zone. Even though the rights to some grandfather durian trees have been individualized and thus benefit individuals or single households and even though the ethic of access has changed somewhat, since most durian fruit is sold rather than consumed (see below), the notion that descent group members retain consumption rights to their ancestral fruit remains strong among village families. Under the ethic of access, an individual may not interrupt the descent group's temporal management zone by cutting a productive tree that an ancestor had planted for "all my children and grandchildren."29 Durian yields can be sporadic but highly profitable. A good durian harvest can earn the tree owner hundreds of thousands of rupiah in one or two months' time.30 In 1991, a bumper crop of durian and other fruit was produced; on one day (February 10) at the end of this season, I counted some 10,000 fruits being sold out of the hamlet (Peluso and Padoch 1996). The following two years were generally poor or mediocre production years, although some individuals with widely scattered fruit holdings had some production; but 1994 was another bumper crop. Of those in the 1991 sample who would venture

28 This state of danger is called kemponan and is commonly accepted throughout Borneo as an outcome of refusing any food, cigarettes, or a betel chew. A person may also be put into the dangerous state of kemponan if they go into someone's house and are not offered such refresh- ment (Robert Sulis Ridu, personal communication, 1996). 29 A rapid survey of six other villages with similar resource management traditions in 1991 (Padoch and Peluso, unpublished data), found in some places that durian trees can only be cut after ritual recognition of the planter and his or her descendants; in others, customary fines must be paid to the village and the descendants of the planter whenever durian is cut. I never came across such practices in Bagak. 30 At the time of research, one U.S. dollar equals approximately 2,000 Rupiah.

This content downloaded from 173.250.174.7 on Wed, 15 Nov 2017 17:38:14 UTC All use subject to http://about.jstor.org/terms 534 NANCY LEE PELUSO estimates of the income from durian they sold, the average income was Rp. 338,200, with one person reporting Rp. 1,000,000 in cash income alone. (Hulled rice cost approximately Rp. 700 per kilogram in the marketplace at the time, compared to a village price of Rp. 300-500 for a single durian.) For some families, these windfalls accounted for as much as one-half the year's cash income.31 Moreover, these sales figures do not account for durian consumed or given away nor for the durian made into tempuyak, a kind of preserved, salted durian. In good durian harvest years, such as 1994, many people reported selling 10 or more kilograms of tempuyak at Rp. 1,000 to 1,500 per kilogram.

Rice and Rubber-A Comparison Rice and rubber are the other important components of Bagak's physical and social landscape. They are also imbued with specific values affecting the bundles of rights associated with them. In former days, self-sufficiency in food production was highly valued, largely of necessity. Prior to the turn of the twentieth century, minimal restrictions on agricultural expansion and the availability of extensive areas of forest for planting hill rice meant that access to land for swiddens was not a constraint. For a lack of cash, a preference for hill rice, and their capacity to be largely self-sufficient in food production at the time, Bagak Dayaks reportedly rarely bought the rice grown in irrigated paddy fields by local Chinese planters. As discussed in the history section above, many factors helped change the nature and role of swidden production in Bagak. Contrary to what might be expected, the acquisition of irrigated rice land left behind by the Chinese was of minor importance at the time. Irrigated rice yields were extremely low in the first twenty years after the villagers took over these lands from the Chi- nese; if anything, people could only produce enough from these fields to feed their families for a couple of months. Many people did not even bother to plant, given the poor yields. Nevertheless, the irrigated fields served as a set of reserve lands for future production-a luxury which used to be common in the form of reserved swidden fallows and forest but which was being curtailed by changes in the regional political economy. After years of bad harvests and the repeated failure of hybrid seeds provided by a government anxious to get swiddeners to plant irrigated rice, people returned to planting traditional varieties (those which the Chinese used to plant) and had very successful harvests between 1991 and 1993. How irrigated fields will be inherited is a question that will have to be answered in the next generation. However, initial trends among the few families in which this land was passed to the next generation indicate variation:

31 Income from other fruits-particularly langsat and angkaham-varied and were not reported systematically.

This content downloaded from 173.250.174.7 on Wed, 15 Nov 2017 17:38:14 UTC All use subject to http://about.jstor.org/terms FRUIT TREES AND FAMILY TREES 535

The land is sometimes treated as a chunk of land allocated to a single child, sometimes reserved for the person who continues the household, and sometimes it is divided between two or more children. Swidden fields were commonly inherited by the descendants of the forest clearer. As with trees, primary rights in swidden fallows went to the child who continued the parents' household. All grandchildren retain some claims on these lands, although the third-generation primary inheritor is the child who continues the household, which is different than the rule of third-generation co-heirs' common inheritance of fruit trees. Similar to other inheritance practices, however, an ethic of access to land played a key role-not surprisingly, given the direct dependence of people on the land for their sustenance. In the case of land for swiddens, the primary rights holder was expected to allow other kin access to land to make swiddens from old fallows which he or she was not actively cultivating. Such land was borrowed in the sense that rent or other payment was rarely if ever paid for the rights of access. Some families would have rules about the length of time that a plot needed to remain in fallow before it could be re-cultivated by themselves or borrower-kin; these rules, however, would also change as the population and other pressures on land changed. Moreover, borrowing land to plant trees was generally not allowed because it would shift the lines of inheritance. In 1990 and 1991, rice from both swiddens and irrigated fields yielded enough to provide a food supply sufficient to feed the people who had produced it for an average of four months. Hill rice yields from swiddens have declined as swidden sizes have shrunk. Only 58 percent of the village sample cultivated swiddens in 1990-91; all who did cultivated one swidden field of an average size of less than 0.40 hectares, as mentioned above. In 1994, 41 percent of the villagers planted swiddens averaging 0.42 hectares. From their swiddens in 1991, villagers reported an average yield of approximately 250 kilograms of paddy (unhulled rice, cleaned only of stems). Since irrigated rice yields have varied greatly and tend to be quite low, many people try not to depend on them. Rubber is the newest component in the local agroforestry system and is the only one of these three major land uses intended specifically and exclusively for sale, since rubber has no local use except when unproductive trees are cut down and used for firewood. Rubber is, however, the primary source of regular income used to purchase food and other subsistence items by nearly every family in the village. When a co-heir of a swidden fallow decides to grow rubber, the land must be taken out of the cropping cycle for the thirty years or so that rubber matures and produces. Before actually planting the rubber, the cultivator generally clears the secondary growth (which can be quite dense even after a few years' fallow) and plants rubber trees in the swidden. Rights to these rubber trees inhere to the planter and his or her household. When the planter dies, any

This content downloaded from 173.250.174.7 on Wed, 15 Nov 2017 17:38:14 UTC All use subject to http://about.jstor.org/terms 536 NANCY LEE PELUSO living trees (or a garden of them), like the swidden fallows of the parent, are passed on to only one of the children. The inheritor may be the child who continues the household or rubber plots may be divided among the children. Rubber rarely lasts through more than two generations, but in cases in which it does, rubber gardens are passed on to individual grandchildren of the planter. It is, therefore, inherited in the same manner as other movable property (harta). Rubber has none of the social or ritual meaning attached to durian. Unlike durian, rubber does not cause the community nor even a whole descent group to form any emotional attachments to it. Neither rubber trees nor rubber gardens are named, and they are not shared by the descent group. Some of this may be attributed to its biological characteristics: It has a relatively short span of production and life, cannot be divided except in arrangements between two parties (owner and tapper) for share tapping, and has no subsistence use. Perhaps because it is the most recent addition to the local landscape, rubber lacks an embedded ethic of access. Ideally, when a rubber garden stops producing, the land should have been returned to the descent group's pool and the rights to clear it-to invest labor-should have been vested in the grandchildren and great-grandchildren of the original forest clearer, just like any swidden fallow. In practice, swidden fallows planted in rubber are effectively privatized. Even if the land were to return to the descent group after the rubber went out of production, the thirty to forty years over which rubber occupied land of the descent group constitutes a much longer time than that taken by rice and vegetables and fallow vegetation. The temporal zoning of land rights is, thus, significantly altered by the planting of rubber. Some 85 percent of sample households owned rubber trees in 1991; and, at the time of the survey, 76 percent regularly collected rubber from them. The average yield per collection day is 6.15 kilograms of crudely processed rubber (sheets). People have different patterns of collecting rubber: Some collect local varieties of rubber on alternate days to allow enough latex to collect in the tree; others do it every day. Some own enough trees to have two routes, collecting from half their trees on one day and the other half on the next. On very rainy or very dry days, people do not tap local rubber. On very rainy days, water overturns the collection cups; and on very hot, dry days, the latex dries up. In 1991, just over a third of the sample owned producing high- yielding trees (High Yielding Varieties, or HYVs); average daily yields from these were approximately 6.75 kilograms.32 The HYVs have been bred to allow for daily tapping-the latex runs even on very dry days.33 People tend to collect rubber year-round, with breaks during the rice harvest and highly

32 Production was allegedly promoted as potentially much higher; for various reasons these goals have not been reached by most participants in the scheme. 33 The impracticality of tapping even these high-yielding varieties of rubber trees in heavy rain usually precludes it.

This content downloaded from 173.250.174.7 on Wed, 15 Nov 2017 17:38:14 UTC All use subject to http://about.jstor.org/terms FRUIT TREES AND FAMILY TREES 537 productive fruit seasons. In a year of unusually good fruit production, rubber generally provides only one-half of the total family cash income from agriculture. In average or poor fruit production years, rubber provides nearly all of it. The biological characteristics of the tree and its wide distribution allow rubber to provide the most regular, year-round source of income to people. It is, very definitely, the most common subsistence crop. This dependence on rubber, combined with an ethic of access earlier associated with access to rice (swidden) land, may explain its importance and popularity in the local landscape. As it became clear that producing rubber was replacing rice self-sufficiency as the main way of providing for a family's subsistence, it would be difficult for a person holding primary rights to swidden fallows to deny a kinsperson in need of access to fallows for growing rubber. Thus, while the latex and the rubber tree have no particular ethic of access attached to them, the land on which the trees are planted does. Rubber is thus, socially, an unique mixture of the property rights and resource characteristics of the rice and fruit tree components of the landscape. Like fruit, rubber lasts a long time and ties up large spatial and temporal zones of production and resource access. But like land planted in rice, the product of land under rubber-latex-is controlled by only one person or household. The ethic of access to it does not entail the division of the production in the same way that fruit is divided among co-heirs. Thus, while the introduction of long-living trees into the swidden system and the landscape had the potential to change the practice and ethics of land inheritance, the control of resources across space and through time has been influenced by the characteristics of the trees, their singular or multiple purposes and meaning, and in all likelihood the type of labor investment required to produce the crop. In sum, managed forests of economic trees have been gradually taking over the swidden fallows of Bagak for some sixty years. Not all the species have equal significance in the economy or the social fabric. I have focused on the two most important species in the discussion above. Rubber owned by the planter (and his or her household) is the most crucial daily subsistence contri- bution to most village households. However, because it rarely stays in the ground through more than two generations, it is planted exclusively for com- merce and has no use value. The ethic of access attached to it is embedded in the land rather than the resource itself. Because it is the most valued and valuable of the fruits, lasts five to seven generations, is locally consumed as well as sold, and has deep social meaning to individuals, direct descendants of the planter, and the community as a whole, durian is perhaps the most important visible evidence of the community's settlement history. Both durian fruit and trees have deep-rooted ethics of access attached to them. Hill rice, the main product of swidden fields, is in decline and may disappear entirely from this landscape except as the occasional interim use between generations of economic trees. It remains to be seen whether the successful harvests of

This content downloaded from 173.250.174.7 on Wed, 15 Nov 2017 17:38:14 UTC All use subject to http://about.jstor.org/terms 538 NANCY LEE PELUSO

irrigated rice between 1991 and 1994 will encourage for generations to come the continuation of current landscape relations-rice in the lowlands and mixed forests of economic trees on the adjacent hills-and the way in which the ethic of access to irrigated rice land will develop.

COMMODIFICATION AND THE INDIVIDUALIZATION OF RIGHTS

IN DURIAN

Selling your durian trees is like selling your own grandfather (Pak Po'on, Bagak Sahwa).

When market demand is as great as it is for durian, when it must be sold immediately after it falls, and when market access is easy, all members of a descent group are concerned when even a small amount of fruit is wasted. The capacity of a co-heir to harvest his or her trees thus takes on increasing importance when the co-heir's group formally or informally allocates access to their trees.34 The market boom in fruit has led to two major changes in durian management. First, many families have permanently allocated panene'an trees to individuals. Second, people are opening up tracts of land-old fallows, old rubber gardens, and sometimes, the edges of the nature reserve- specifically to plant durian and other fruit gardens and thereby to claim the land for subsequent generations. Nevertheless, the trend towards individualization of durian rights is not absolute. Households and extended families vary in the types of tree tenure arrangements that they maintain, and some village trees are still managed for rotation among descent group rights holders or for open access. As mentioned above, social and ecological factors shape the distribution of access to descent group trees. By consensus or by default, descent groups and the person in charge of their common trees determine which particular combi- nations of inheritance rights and labor investments will govern the allocation of panene'an trees that they hold together. Four considerations seem to weigh heavily in this decision: first, the number of trees which the descent group holds; second, the relative yields of various trees; third, the productivity of each heir's allocated trees in previous years; and, fourth, the other resources to which each of the co-heirs has access. As before, some interpretation of the ethic of access mediates the allocation decision. Families who use a relatively structured means of allocating access (for example, by meeting at the beginning of a season) face ecological and social obstacles to their decision making. At the time of year when access rights are

34 In former days, although much of the durian that could not be consumed was made into tempuyak or lempok, some was sold or kept for later consumption. In 1991, I found only a few people who had made tempuyak; this changed in the bumper harvest of 1995, when a majority of the sample made and sold approximately 10 kilograms of tempuyak per household (Simon Takdir, personal communication).

This content downloaded from 173.250.174.7 on Wed, 15 Nov 2017 17:38:14 UTC All use subject to http://about.jstor.org/terms FRUIT TREES AND FAMILY TREES 539 discussed, many trees' actual yields are unknown. Actual yields may be adversely affected by unforeseen circumstances, such as heavy winds or rain, which may blow off abundant flowers. As a result, durian yields do fluctuate each year, although some trees produce nearly every year. Usually older trees with many branches produce a great deal of fruit, but at some point their productivity wanes significantly. Some years nearly the whole village has highly productive harvests; others produce nothing for anyone; still other years offer bumper harvests in mountain gardens and nothing in the flatlands. Even within the same section of village forest, variation is common. Adjacent trees may differ significantly in productivity, depending both on the site's characteristics and the owners' management. Despite attempts by families to balance access to the fruit, some villagers find they are out of luck and have no access to productive, inherited trees in certain years, although another alternative-leasing trees-can be used.35 Moreover, the ethic of access will enable them to join a relative's harvest in order to have some fruit for consumption. The changes in durian management are apparent in both temporal and spatial ownership patterns. Temporal changes in inter-generational rights to durian are most evident in the distribution of third-generation or older trees because, ideally, it is in that generation that a large descent group holds the rights. In 1991, more than one-third of the sample families reported having private rights to panene'an trees, citing reasons such as grandparents and parents wanting to avoid disputes among children and grandchildren. Discussions with sample villagers in 1996 indicated that 68 percent had privatized their panene'an trees, but only 42 percent of these held the private rights. Only 12 percent of the sample said that they had not privatized any panene'an trees. Individualization is also affecting spatial zoning of rights, imparting value to land; but value in the spatial territory is a function of the number, productivity, and value of the trees planted there. Long-living trees planted in swidden fallows tie up the land for much longer than the average unimproved fallow. Increasingly, people clear fallows with the intention of planting some kind of forest garden immediately after the rice crop: Some 47 percent of those who planted swiddens in 1993, and 42 percent of those who did so in 1994, planted trees in the fallows immediately after the harvest. Rubber, with its numerous trees planted in proximity to facilitate latex collection, secures spatial zones for an individual or a household. Formerly planted intermittently next to other

35 A discussion of leasing arrangements and their implications for individual versus group resource control is beyond the scope of this essay. Contrary to the experience in many agrarian contexts, however, leasing out does not occur only in times of hardship but is often an option chosen by a family when it does not have access to sufficient labor to harvest all their fruit. This is particularly true in years of good rice yields and good fruit yields-when the harvesting seasons for each coincide. Similarly, the need to lease trees can be caused not only by a surplus of wealth but also by a lack of fruit from one's own trees in a particular year.

This content downloaded from 173.250.174.7 on Wed, 15 Nov 2017 17:38:14 UTC All use subject to http://about.jstor.org/terms 540 NANCY LEE PELUSO people's individual trees, durian tied up small spaces for multiple generations of descendants-a temporal rather than spatial zoning of private rights. How- ever, as durian trees begin to be planted in places they were never planted before, in large clusters and in swidden fallows, they promise to expand their influence in spatial terms. The individualization of descent group rights to grandparent durian trees occurs most commonly in one of three ways: The descent group may agree to distribute its commonly held trees among individual members; a grandparent may designate specific trees or clusters of trees to individual grandchildren; or individuals may usurp the ownership group's authority and assume possession of a grandfather tree, usually by planting young trees around mature descent group trees or simply by building a shelter there year after year without consulting the group. By planting trees around a panene'an tree, once the younger trees come into production, the individual takes over the harvest of the group's tree. This is essentially a territorialization strategy-a squatting by one individual or branch of the family on the trees held under the common (temporal) rights of the larger descent group. The usurpation of rights takes place with the descent group's implicit consent, since all the tree planter's descendants have rightful claims to the tree's fruit. In several cases where this had happened, other family members explained the group's failure to exert their collective claims as an effort to avoid public discord. It may also be seen as a new interpretation of the ethic of access, motivated by the trend toward individualization and territorialization of rights. Individualization of descent group rights is by no means universal. Some families insist that the grandchildren and great-grandchildren of a tree planter are violating important customary practices in privatizing access to grandparent trees (Peluso and Padoch 1996). Indeed, when I first started asking about the privatization of panene'an trees, several of my key informants insisted I was wrong. They refused to believe it could be happening until they heard it themselves from the individuals concerned. This sparked a great deal of discussion amongst those who felt it was not an appropriate allocation of resources. Nevertheless, others insisted they had not and would not individualize panene'an trees because they were meant to be owned jointly and to remain so. Whether or not families have individualized the rights to their grandparent trees, nearly all villagers express the feeling that individually controlled trees are easier to manage and potential disputes are avoided, while the fruits of descent group trees serve to support those in the family whose own trees do not produce yet. This preference for individually held resources was part of the reason that a person would plant his or her own trees. The other was the felt ethical obligation to take oneself out of the descent group's pool of resources as soon as possible. Panene'an had served two types of redistributive purposes in the past and, among the families following the old practices, still do. The first type is a

This content downloaded from 173.250.174.7 on Wed, 15 Nov 2017 17:38:14 UTC All use subject to http://about.jstor.org/terms FRUIT TREES AND FAMILY TREES 541 redistribution of village resources between age groups. Young couples always depended on panene'an until their parents' or their own durian trees bore fruit, at which time many of them stopped participating in the harvest of panene'an trees. The second type is a redistribution which aids the weaker, less diligent, or less clever farmers, who benefit from continued access to the trees and fruit of their ancestors with little effort. Most villagers regard long-term dependence on the trees planted by one's forebears as both a sign of laziness and a handicap to one's children and grandchildren. Unless today's farmer plants fruit trees, his or her children will suffer. Thus, a lifetime of dependence on panene'an trees lowers one's general regard in the eyes of other members of the community and constrains one's descendants' ability to prosper from the land. What aspects of commodification have most changed the landscape and the ethics of access to durian, one of its key components? Durian has long been a commodity and has long been sold by Bagak villagers. The commodification of durian began before the paving of the road and the growth of nearby cities. In some ways, rubber began the major landscape transformation from one dominated by fields and fallows to one dominated by producing, economic forests, even though durian is one of the oldest cultivated trees in the landscape, if not the oldest. The ecology and biology of rubber-particularly its capacity to produce year-round for a market which had fluctuating prices but was nonetheless available-gave it an edge over durian as a means of providing cash for the family's subsistence when sedentarization and political- economic change began to cause the decline in rice's viability as a subsistence crop. Durian harvests, if they occurred at all, were limited to one time of the year; and the fruit could not be stored except as tempuyak, a product of much lower value. Once rubber had provided a strong subsistence base for virtually the whole village, other changes in the regional political ecology spurred the spread of durian and its transformation to a consumption and a commercial crop with multiple meanings and values. Now the burgeoning urban markets for fruit, the transportation infrastructure to which the village had direct access, the propensity of their ancestors to plant durian even in the forbidden swidden fallows within the nature reserve, and the perseverance (and luck) of the village leaders who refused to contract village land to encroaching plantations on their borders, have put Bagak durian at a premium and led to people planting it in many more unfamiliar sorts of places. In the current historical moment, durian is becoming spatialized, both in terms of its occurrence and its control. How this will affect the ethic of access to durian fruit, trees, and the land on which they are planted will be an outcome of the future process.

CONCLUSION

In this essay, I have examined changes in the landscape of a village in West Kalimantan and the interactions of environmental change with resource tenure

This content downloaded from 173.250.174.7 on Wed, 15 Nov 2017 17:38:14 UTC All use subject to http://about.jstor.org/terms 542 NANCY LEE PELUSO

patterns. The case is instructive on several points. First, in terms of understanding landscape processes, the experience of Bagak confounds the stereo- typical understanding of the transformation and agricultural intensification of a tropical people's forest and land, that is, an understanding predicated on an evolutionary model that starts with what is assumed to be a natural forest and moves toward the production of field crops for food. In such cases, intensification of agricultural production is seen to involve increasing labor or capital inputs per unit of land and an emphasis on commercial or subsistence field agriculture, usually in concert with increases in population growth and sedentarization (for example, Geertz 1963; Boserup 1981; Dove 1985). In contrast to this pattern, villagers acting in this Kalimantan case are moving away from field and food crop agriculture and toward the greater production of forests and agroforestry products. Over the past three generations, villagers have altered their hillside vegetation from a mixture dominated by swidden fields and fallows with patches of managed forests to a heavily managed forest landscape dominated by selected useful trees. The villagers' increasingly intensive tree planting on village lands was stimulated not only by land scarcity engendered by enforced sedentarization and by the loss of their formal access (legal rights) to ancestral land when the colonial government carved a nature reserve from the villagers' land but also by the increased market access that resulted from road improvements, urbanization, population growth, and sedentarization. Moreover, their contemporary management activities build on traditional procedures, preferences and property rights for forest management procedures, though both the selection of species and the balance between private and group rights to the resources may have changed. The case strongly sug- gests that fruit and forest product production-not unmanaged extraction- has been an important element of the resource management strategies of Bornean people for hundreds of years. Further, the production of specific products has necessitated a form of ecosystem management or the manipulation- sometimes by tacit agreement among villagers, sometimes through conflict- of the field and forest components of the landscape. The study, therefore, puts a different slant on much of the literature on "extractive economies," which tends to underplay the purposive management roles that local people take (see, for example, Bunker 1985; cases in Nepstad and Schwartzman 1992; Dove 1995; but compare, Hecht, Anderson, and May 1988; Posey and Hecht 1985; Padoch 1994). The second surprise of this case has to do with the unintended outcomes and unanticipated impacts of government intervention on locally managed landscapes. Government land management policies, sedentarization, changes in regional markets, and national political events initially caused multiple hardships for Bagak by creating new spatial land-use zones and redefining legal versus illegal land uses and access rights. Often such interruptions of

This content downloaded from 173.250.174.7 on Wed, 15 Nov 2017 17:38:14 UTC All use subject to http://about.jstor.org/terms FRUIT TREES AND FAMILY TREES 543 local management systems cause widespread degradation and poverty, but local people created a new and well-managed social and ecological landscape by building on economic components of their earlier land use system (mixed fruit forests), adding other economic components (rubber gardens), rein- terpreting access rights to the nature reserve (creating a buffer zone of low- impact use), by maintaining and altering their ethics of access to various ecosystem components in response to political-economic change, and by taking advantage of the relocation of some of their neighbors and kin. The new forested landscape and the buffer zone of useful and managed trees planted within the government-claimed territory of the reserve can be seen as physical evidence of local resistance to, and counter-appropriation of, government appropriation of management and property rights. Every generation of villagers has come together in different ways to contest the boundaries of the government nature reserve, thus placing the ethic of access at the very heart of state and local struggles over resource control. In yet another context, then- one which was significantly altered by government intervention-the Bagak landscape serves as a living record of village history. The case thus illustrates the need to understand different versions of environmental and social history in order to be able to read a landscape. This leads to a third, less clear but suggestive, conclusion for this case that has to do with ethnicity and landscape construction and management. Forced by the Dutch to abandon their "traditional" landscape, the Salako had to alter significantly their way of interacting with that landscape in several dimensions. They moved from living arrangements in groups (longhouses) to single-family homes, albeit ones large enough to accommodate some extended family members. Similarly, villagers were slowly pushed by circumstances to reconceptualize their resource rights and the spatial arrangement of their landscape-particularly its extent-and to move toward a territorialization of these rights from an emphasis on rights in trees or crops. As mentioned below, however, their ethic of access has somewhat tempered the trend toward individualization and territorialization of all rights to all resources. What has not been emphasized in this analysis but emerges as a possible theme is the notion that these new landscape arrangements and rights to some degree reflect those practiced by the Chinese as early as the nineteenth century. Despite the question of whether Chinese practices were a result of, or just similar to, Dutch conceptions of appropriate spatial arrangements for resource management, the two methods of creating spatial zones were already concur- rent. In the Indonesian era, the government's preferred land management style more closely reflects the spatial landscape arrangements of its Chinese and Dutch predecessors than it does those of the local traditional systems. Some impact on the way in which local people conceptualize and practice their management strategies has been inevitable. Indeed, one could argue, perhaps, that the impact of these different landscape conceptions on the consciousness

This content downloaded from 173.250.174.7 on Wed, 15 Nov 2017 17:38:14 UTC All use subject to http://about.jstor.org/terms 544 NANCY LEE PELUSO

of Salako people was strong enough to cause them to participate in evicting the Chinese from their farms in the 1960s. In other words, the political upheaval of that period was not simply about resource scarcity or an anti- communist popular uprising but one violent outcome of the remaking of people's vision of their landscape. A fourth point can be made about the relationship between biology, economic behavior, property rights, and culture. It is not unusual to recognize that prices, policies, and legal systems affect formal property rights. I have argued that a tree's biological characteristics also influence property rights. More important, however, the durian trees' social value extends beyond economics, biology, and even the claims of the planter's descendants because the entire village views the trees as markers of settlement history and territorial claim. This case shows how a historically grounded set of meanings attached to a resource can critically mediate both exclusionary and incentive-based rights structures. We saw first how a system of property rights in trees increased the number of rights-holders with each succeeding generation. Al- though I have also shown that some rights holders had greater control over these resources than others, the notion that all descendants should retain some access rights contradicts a stereotyped model predicting a linear procession of property relations from open access to common property to private property (Bromley 1989). What I have called the ethic of access is key among the reasons why broader societal trends and government encouragement of privatization and individualization of resource access have not unilaterally transformed the property rights in trees. This ethic has enhanced and been enhanced by kinship ties and community investments in the social meanings of past and present landscapes. As a result, although intensification of forest production and rapid improvement of market access exerts strong pressure to make resource tenure private or individualized, the social meanings of these resources and their links to the collective and individual social histories of villagers have caused many people to resist the divisive aspects of commodification. We have also seen that, although the ethic of access is gener- ically discussed, it seems to have somewhat less application toward the goal of leveling female and male control of key agrarian resources and, more obviously, in negotiating rights across ethnic groups under conditions of rapid change. In the case of the agrarian unrest of the 1960s and 1970s, a new ethic of access had already been brewing in which the rights of resource access were being seen in ethnic terms. In this study, I have shown that the notion of zoning has both temporal and spatial aspects. These aspects are best illustrated by the analysis of property rights in long-living trees but can also be understood or analyzed through an examination of the set of resource tenure arrangements that people are engaged in at different points in their life cycles. The ethic of access worked to regulate this temporal zoning as well, lending credence through its discourse

This content downloaded from 173.250.174.7 on Wed, 15 Nov 2017 17:38:14 UTC All use subject to http://about.jstor.org/terms FRUIT TREES AND FAMILY TREES 545 to the redistribution of access between people of various age groups. The ethic suggested not only at which point in life one has more widely recognized claims to resources held in common than those of other group members but also at which point people are expected to stop exercising their ancestral claims. How ethics of access are transformed in the future will depend to a large degree on whether and how temporal zones of rights give way to the spatialization of resource control. Finally, I have shown that the study of resource access and landscape as social processes interacting with ecological factors is much more complex conceptually than thinking of property and landscapes as inanimate objects. If landscape is thought of as an artifact of human consciousness and therefore subject to multiple interpretations, visions, and memories (Schama 1995), it becomes ever more important to understand how the meanings and value of that landscape shape the processes and institutions of access to it.

REFERENCES

Appell, George. 1970. "Comparisons of Land Tenure Systems among the Dayak of Borneo." Unpublished manuscript. . 1971. "Systems of Land Tenure in Borneo: A Problem in Ecological Deter- minism." Borneo Research Bulletin, 3:17-21. Alcorn, Janis. 1981. "Huastec Noncrop Resource Management: Implications for Pre- historic Rainforest Management." Human Ecology, 9:395-417. Aschmann, Homer. 1988 [1963]. "Proprietary Rights to Fruit on Trees Growing on Residential Land," in Whose Trees? Proprietary Dimensions of Forestry, Louise P. Fortmann and John W. Bruce, eds., 63-68. Boulder, CO: Westview Press. Balee, William. 1989. "The Culture of Amazonian Forests," in Resource Management in Amazonia: Indigenous and Folk Strategies, D. A. Posey and W. Balee, eds., vol. 7, 1-21, of Advances in Economic Botany. New York: New York Botanical Garden. Berry, Sara. 1989. "Social Institutions and Access to Resources." Africa, 59:1,41-55. Blaikie, Piers. 1985. The Political Economy of Soil Erosion in Developing Countries. London: Longman. Blaikie, P.; and H. Brookfield. 1987. Land Degradation and Society. London: Methuen. Boserup, Esther. 1981. Population and Technology. Oxford: Blackwell. Bromley, Daniel W. 1989. "Property Relations and Economic Development: The Other Land Reform." World Development, 17:6, 867-76. .1991. Environment and Economy: Property Rights and Public Policy. Cam- bridge: Basil Blackwell. Brosius, Peter. n.d. "Landscape and Society in Borneo: Penan Hunter-Gatherers of Sarawak." Unpublished manuscript. Bryant, Raymond. 1992. "Political Ecology: An Emerging Research Agenda in Third- World Studies." Political Geography, 11:1, 12-36. Bunker, Stephen. 1985. Underdeveloping the Amazon. Chicago: University of Illinois Press. Chew, Daniel. 1990. Chinese Pioneers on the Sarawak Frontier. Singapore: Oxford University Press. Colfer, Carol; with Richard G. Dudley, and in cooperation with Herri Hadikusumah, Rusydi, Niken Sakuntaladewi, and Amblani. 1993. Shifting Cultivators of Indo-

This content downloaded from 173.250.174.7 on Wed, 15 Nov 2017 17:38:14 UTC All use subject to http://about.jstor.org/terms 546 NANCY LEE PELUSO

nesia: Marauders or Managers of the Forest? Rome: FAO Community Forestry Case Study Series, No. 6. Colfer, Carol; and Herwasono Soedjito. 1995. "Food, Forests and Fields in a Bornean Rainforest: Toward Appropriate Agroforestry Development," in Borneo in Transi- tion: People, Forests, Conservation, and Development, Christine Padoch and Nancy Lee Peluso, eds. Kuala Lumpur: Oxford University Press. Denevan, William M.; and Christine Padoch, eds. 1987. Swidden Fallow Agroforestry in Latin America. Bronx, NY: The New York Botanical Garden: Advances in Economic Botany (7). Dove, Michael R. 1985. "The Agroecological Mythology of the Javanese and the Political Economy of Indonesia." Indonesia, no. 39, 1-36. .1986. "Peasant Versus Government Perception and Use of the Environment: A Case Study of Banjarese Ecology and River Basin Development in South Kali- mantan." Journal of Southeast Asian Studies, 17:3, 113-36. . 1988. The Real and Imagined Role of Culture in Development. Honolulu: University of Hawaii Press. . 1993. "Rubber Eating Rice, Rice Eating Rubber." Paper presented to the Agrarian Studies Seminar, Yale University, September, 1993. .1996. "So Far from Power, So Near to the Forest: A Structural Analysis of Gain and Blame in Tropical Forest Development," in Borneo in Transition: People, Forests, Conservation, and Development. Christine Padoch and Nancy Lee Peluso, eds. Kuala Lumpur: Oxford University Press. Ex, Juron. 1992. "Report on Social Forestry among Villagers in Sanggau." Pontianak: GTZ. Unpublished manuscript. Fortmann, Louise P. 1985. "The Tree Tenure Factor in Agroforestry with Particular Reference to Africa." Agroforestry Systems, 2:229-51. . 1990. "Locality and Custom: Non-Aboriginal Claims to Customary Usufruc- tuary Rights as a Source of Rural Protest." Journal of Rural Studies, 6:2, 195-208. Fortmann, Louise P.; and John W. Bruce, eds. 1988. Whose Trees? Proprietary Di- mensions of Forestry. Boulder, CO: Westview Press. Freeman, J. D. 1955. Iban Agriculture: A Report on the Shifting Cultivation of Hill Rice by the Iban of Sarawak. London: HMSO. Geddes, William. 1954. Report on the Land Dayaks of the First Division. Kuching: Colonial Printing Office. Geertz, Clifford. 1963. Agricultural Involution. Berkeley: University of California Press. Hecht, Susanna; Anthony Anderson; and Peter May. 1988. "The Subsidy from Nature: Shifting Cultivation, Successional Palm Forests, and Rural Development." Human Organization, 47:25-35. Hecht, Susanna; and Alexander Cockbur. 1989. The Fate of the Forest: Developers, Destroyers, and Defenders of the Amazon. London: Verso. Lahjie, Abubakar M.; and Bertold Seibert. 1988. "Agroforestry: Untuk Pembangunan daerah Pedesaan di Kalimantan Timur." Proceedings of a seminar held September 19-22, 1988, by Fakultas Kehutanan, Universitas Mulawarman, Samarinda and GTZ, German Forestry Group. Leach, T. A. 1988 [1919]. "Date-Trees in Halfa Province," in Whose Trees? Propri- etary Dimensions of Forestry, Louise P. Fortmann and John W. Bruce, eds., 43-48. Boulder, CO: Westview. Leach, Melissa. 1995. Rainforest Relations: Gender and Resource Management among the Gola of Sierra Leone. Washington, D.C.: Smithsonian Press. Leaman, Danna J.; and Razali Yusuf; and Harini Sangat-Roemantyo. 1991. Kenya

This content downloaded from 173.250.174.7 on Wed, 15 Nov 2017 17:38:14 UTC All use subject to http://about.jstor.org/terms FRUIT TREES AND FAMILY TREES 547

Dayak Forest Medicines. Jakarta: World Wide Fund for Nature/Indonesia Pro- gramme. Michon, G; and F. Mary Michon. 1994. "Conversion of Traditional Village Gardens and New Economic Strategies of Rural Households in the Area of Bogor, Indo- nesia." Agroforestry Systems, 25:1, 31-58. Moore, Sally Falk. 1986. Social Facts and Fabrications: "Customary" Law on Kilimanjaro, 1880-1980. Cambridge: Cambridge University Press. Nepstad, Daniel; and Stephen Schwartzman. 1992. "Non-Timber Forest Products as Sustainable Management Strategies." Advances in Economic Botany, 9:3. Bronx, NY: The New York Botanical Garden. Neumann, R. 1992. "The Political Ecology of Wildlife Conservation in the Mount Meru Area of Tanzania." Land Degradation and Rehabilitation, 3:85-98. Oakerson, Ronald J. 1986. "A Model for Analysis of Common Property Problems," in the Proceedings of the Conference on Common Property Management, Panel on Common Property, ed., 3-30. Washington, D.C.: National Academy of Sciences. Obi, Chinwuba. 1963. The Ibo Law of Property. London: Butterworths. Okoth-Ogendo, H. W. 0. 1989. "Some Issues of Theory in the Study of Tenure Relations in African Agriculture." Africa, 59:1, 6-17. Ostrom, Elinor. 1990. Governing the Commons: The Evolution of Institutions for Collective Action. New York: Cambridge University Press. Padoch, Christine. 1983. Migration and its Alternatives Among the Iban of Sarawak. The Hague: Martinus Nijhoff. .1994. "The Woodlands of Tae: Traditional Forest Management in Kaliman- tan," in Forest Resources and Wood-Based Biomass Energy as Rural Development Assets. New Delhi: Oxford and IBH Publishing Company, Ltd. Padoch, Christine; and Charles Peters. 1993. "Managed Forests of West Kalimantan, Indonesia," 167-76, in Perspectives on Biodiversity: Case Studies of Genetic Re- sourcesfor Conservation and Development. Washington, D.C.: American Associa- tion for the Advancement of Science. Peluso, Nancy Lee. 1992a. "The Ironwood Problem: (Mis)-Management and Develop- ment of an Extractive Rainforest Product." Conservation Biology, 6:2, 210-19. .1992b. Rich Forests, Poor People: Resource Control and Resistance in Java. Berkeley: University of California Press. .1992c. "The Political Ecology of Extraction and Extractive Reserves in East Kalimantan, Indonesia." Development and Change, 23:4, 49-74. .1993. "Coercing Conservation: The Politics of State Resource Control." Global Environmental Change, 4:2, 199-218. Peluso, Nancy Lee; and Christine Padoch. 1996. "Changing Resource Rights in Man- aged Forests of West Kalimantan," in Borneo in Transition: People, Forests, Con- servation, and Development, Christine Padoch and Nancy Lee Peluso, eds. Kuala Lumpur: Oxford University Press. Peters, Charles. 1994. A Practitioner's Guide to Tropical Forest Ecology. Washington, D.C.: World Wide Fund for Nature (WWF) Biodiversity Support Network. Peters, Pauline. 1992. "Manoeuvres and Debates in the Interpretation of Land Rights in Botswana." Africa, 62:3, 413-34. Posey, Darrell. 1985. "Indigenous Forest Management of Tropical Forest Systems: The Case of the Kayapo Indians of the Brazilian Amazon." Agroforestry Systems, 3:2, 139-58. . 1989. "Forest Islands," in People of the Tropical Rainforest, Julie Denslow and Christine Padoch, eds. Berkeley: University of California Press. Potter, Leslie. 1987. "Banjarese Swidden Cultivators in South Kalimantan," in Land

This content downloaded from 173.250.174.7 on Wed, 15 Nov 2017 17:38:14 UTC All use subject to http://about.jstor.org/terms 548 NANCY LEE PELUSO

Degradation and Society, Piers Blaikie and Harold Brookfield, eds. London: Me- thuen. Rousseau, Jerome. 1990. Central Borneo: Ethnic Identity and Social Life in a Strati- fied Society. Oxford: Clarendon Press. Salafsky, N.; and 0. L. Dugelby; and J. W. Terborgh. 1993. "Can Extractive Reserves Save the Rainforest? An Ecological and Socio-Economic Comparison of Non- Timber Forest Product Extraction Systems in Peten, Guatemala and West Kaliman- tan, Indonesia." Conservation Biology, 7:1, 39-52. Sather, Clifford. 1990. "Trees and Tree Tenure in Paku Iban Society: The Management of Secondary Forest Resources in a Long-Established Iban Community." Borneo Review, 1:1, 16-40. Schama, Simon. 1995. Landscape and Memory. London: Harper Collins Publishers. Schroeder, Richard A. 1993. "Shady Practice: Gender and the Political Ecology of Resource Stabilization in Gambian Garden/Orchards." Economic Geography, 69:4, 349-65. Scott, James C. 1976. The Moral Economy of the Peasant: Rebellion and Subsistence in Southeast Asia. New Haven: Yale University Press. Shipton, Parker; and Mitzi Goheen. 1992. "Understanding African Landholding: Pow- er, Wealth, and Meaning." Africa, 62:3, 307-27. Smith, Neil. 1984. Uneven Development: Nature, Capital and the Production of Space. Oxford: Basil Blackwell. Weinstock, J. A. 1983. "Rattan: Ecological Balance in a Borneo Rainforest Swidden." Economic Botany, 37:1, 58-68. Wolf, Eric. 1982. Europe and the People Without History. Berkeley: University of California.

This content downloaded from 173.250.174.7 on Wed, 15 Nov 2017 17:38:14 UTC All use subject to http://about.jstor.org/terms